An electrical device, in particular a data recorder incorporating non-volatile memory, that survives and operates during the effects of physical shocks.
Recording data from tests of ordnance systems employed in a high shock environment, such as target penetrators, dual-stage warheads, etc., presents significant practical problems in acquisition and retention of test data. Possible conventional solutions include:
a. Telemetry through direct instrumentation, i.e., wiring directly to the device under test. Constraints of this method are that the system under test must be stationary and that the instrumentation cabling must be capable of surviving a high shock environment.
b. Data transceivers using radio or telemetry techniques to transmit environmental data. Constraints of this method are that the transceivers are not available as sufficiently shock-hardened units for high shock environments and require a large volume within the test setup.
c. Shock-hardened solid-state data recorders incorporating volatile memory or battery backup. The constraint is that reliability is compromised by power outages or failure of batteries, or both, during and immediately after the test.
One solution involves shock hardening solid-state digital data recorders incorporating non-volatile memories, such as an embodiment of the present invention.
The shock-hardened data recorder of a preferred embodiment of the present invention has five functional elements:
a. a signal conditioning circuit,
b. an oscillator,
c. a 16-bit analog-to-digital converter (ADC),
d. a field programmable gate array (FPGA), and
e. a non-volatile memory.
The signal conditioning circuit incorporates a fixed-output voltage regulator to supply the remaining functional elements. The output is filtered by two capacitors, one on the input side of the regulator and the other on the output side. A surge-suppressing diode eliminates voltage spikes on the output of the supply source.
The oscillator is a shock-hardened oscillator. It supplies a clock signal for use by the FPGA.
The 16-bit ADC converts the incoming analog signal into binary.
The FPGA provides timing and control between the ADC and the non-volatile memory.
The non-volatile reprogrammable memory incorporates both a fast, static random access memory (SRAM) and electrically erasable programmable read only memory (EEPROM). Data within the SRAM is written to the EEPROM either on command or automatically during power down. The SRAM contains a voltage-sensing element that detects when the power supply has dropped below a minimum threshold. Upon breaching this minimum, the SRAM initiates a xe2x80x9cstorexe2x80x9d command. A capacitor from the signal conditioning circuit provides the necessary power to write the contents of the SRAM to the EEPROM during power down. Data stored in the EEPROM is automatically recalled during power up.
At power up, the FPGA determines its status, i.e., its current mode of operation, by interrogation of the trigger input. In xe2x80x9cmemory readxe2x80x9d mode, the FPGA reads the data stored in the non-volatile memory (EEPROM) and forwards the data in a serial bit stream. A logic circuit then outputs the serial bit stream as 16-bit words that represent the original analog signal.
In xe2x80x9cmemory writexe2x80x9d mode, the FPGA enables the ADC and waits for a trigger line to change states. Once the trigger is detected, the ADC is commanded to convert the analog signal into a digital word, nominally 16 bits. The 16-bit word is clocked out of the ADC sequentially as two 8-bit bytes. The most significant byte is loaded into the FPGA first. The most significant byte and memory address are sent to a first memory chip for storage. The least significant byte is loaded next. The least significant byte and memory address are sent to a second memory chip for storage. Thus, one chip retains the most significant byte and the other the least significant byte. The memory address is incremented, and the process is repeated until memory is full. The benefit of this setup is that only four wires need be used to store and retrieve data from the recorder: ground, power, trigger, and serial data out. This significantly reduces the physical protection required and, in turn, facilitates the shock hardening of the recorder.
Advantages of preferred embodiments of the present invention include:
reduced power requirements;
simpler components;
no moving parts;
high reliability due to controlled shutdown should power fail;
reduced system capital costs by eliminating separate power source/batteries;
reduced testing cost by reducing the need to repeat testing where data is lost;
increased operational readiness;
reduced maintenance costs of a simpler design;
high durability;
easy upgrade; and
ready application to existing instrumentation systems.
Embodiments of the present invention also can be applied to test environments with few environmental constraints, resulting in very low mean time between failure (MTBF) rates. Preferred embodiments are fully disclosed below, albeit without placing limitations thereon.